Recently, the high-performance WAN encapsulation method known as Frame Relay has become one of the most popular technologies in use. It operates at the Physical and Data Link layers of the OSI reference model and was originally designed for use across Integrated Services Digital Network (ISDN) interfaces. But today, Frame Relay is used over a variety of other network interfaces. Cisco Frame Relay supports the following protocols:
- IP
- DECnet
- AppleTalk
- Xerox Network Service (XNS)
- Novell IPX
- Connectionless Network Service (CLNS)
- International Organization for Standards (ISO)
- Banyan Vines
- Transparent bridging
Frame Relay provides a communications interface between DTE (Data Terminal Equipment) and DCE (Data Circuit-Terminating Equipment, such as packet switches) devices. DTE consists of terminals, PCs, routers, and bridges—customer-owned end-node and internetworking devices. DCE consists of carrier-owned internetworking devices. Popular opinion maintains that Frame Relay is more efficient and faster than X.25 because it assumes error checking will be done through higherlayer protocols and application services.
Frame Relay provides connection-oriented, Data Link layer communication via virtual circuits just as X.25 does. These virtual circuits are logical connections created between two DTEs across a packet-switched network, which is identified by a DLCI, or Data Link Connection Identifier. (We’ll get to DLCIs in a bit.) Also, like X.25, Frame Relay uses both PVCs (Permanent Virtual Circuits) and SVCs (Switched Virtual Circuits), although most Frame Relay networks use only PVCs. This virtual circuit provides the complete path to the destination network prior to the sending of the first frame.
Frame Relay Terminology:
To understand the terminology used in Frame Relay networks, first you need to know how the technology works. Figure 10.4 is labeled with the various terms used to describe different parts of a Frame Relay network.
The basic idea behind Frame Relay networks is to allow users to communicate between two DTE devices through DCE devices. The users should not see the difference between connecting to and gathering resources from a local server and a server at a remote site connected with Frame Relay. Chances are this connection will be slower than a 10Mbps Ethernet LAN, but the physical difference in the connection should be transparent to the user.
Figure 4 illustrates everything that must happen in order for two DTE devices to communicate. Here is how the process works:
1. The user’s network device sends a frame out on the local network. The hardware address of the router (default gateway) will be in the header of the frame.
2. The router picks up the frame, extracts the packet, and discards the frame. It then looks at the destination IP address within the packet and checks to see whether it knows how to get to the destination network by looking in the routing table.
3. The router then forwards the data out the interface that it thinks can find the remote network. (If it can’t find the network in its routing table, it will discard the packet.) Because this will be a serial interface encapsulated with Frame Relay, the router puts the packet onto the Frame Relay network encapsulated within a Frame Relay frame. It will add the DLCI number associated with the serial interface. DLCIs identify the virtual circuit (PVC or SVC) to the routers and provider’s switches participating in the Frame Relay network.
4. The channel service unit/data service unit (CSU/DSU) receives the digital signal and encodes it into the type of digital signaling that the switch at the Packet Switch Exchange (PSE) can understand. The PSE receives the digital signal and extracts the 1s and 0s from the line.
5. The CSU/DSU is connected to a demarcation (demarc) installed by the service provider, and its location is the service provider’s first point of responsibility (last point on the receiving end). The demarc is typically just an RJ-45 jack installed close to the router and CSU/DSU.
6. The demarc is typically a twisted-pair cable that connects to the local loop. The local loop connects to the closest central office (CO), sometimes called a point of presence (POP). The local loop can connect using various physical mediums, but twisted-pair or fiber is very common.
7. The CO receives the frame and sends it through the Frame Relay “cloud” to its destination. This cloud can be dozens of switching offices—or more! It looks for the destination IP address and DLCI number. It typically can find the DLCI number of the remote device or router by looking up an IP-to-DLCI mapping. Frame Relay mappings are usually created statically by the service provider, but they can be created dynamically using the Inverse ARP (IARP) protocol. Remember that before data is sent through the cloud, the virtual circuit is created from end to end.
8. Once the frame reaches the switching office closest to the destination office, it is sent through the local loop. The frame is received at the demarc and then is sent to the CSU/DSU. Finally, the router extracts the packet, or datagram, from the frame and puts the packet in a new LAN frame to be delivered to the destination host. The frame on the LAN will have the final destination hardware address in the header. This was found in the router’s ARP cache, or an ARP broadcast was performed.
The user and server do not need to know, nor should they know, everything that happens as the frame makes its way across the Frame Relay network. The remote server should be as easy to use as a locally connected resource.
Frame Relay Encapsulation: When configuring Frame Relay on Cisco routers, you need to specify it as an encapsulation on serial interfaces.
There are only two encapsulation types: Cisco and IETF (Internet Engineering Task Force). The following router output shows the two different encapsulation methods when choosing Frame Relay on your Cisco router:
RouterA(config)#int s0
RouterA(config-if)#encapsulation frame-relay ?
ietf Use RFC1490 encapsulation
The default encapsulation is Cisco unless you manually type in IETF, and Cisco is the type used when connecting two Cisco devices. You’d opt for the IETF-type encapsulation if you needed to connect a Cisco device to a non-Cisco device with Frame Relay. So before choosing an encapsulation type, check with your ISP and find out which one they use. (If they don’t know, hook up with a different ISP!)